Also guard against overflow when subtracting from unsigned.

Also when summing multiple hist_entry.count use uint64_t for accumulator.

It's just a simple index.

Addresses Coverity issue 1500290
Signed-off-by: Stephen Hutchinson <qyot27@gmail.com>

Since the check got simplified and stdbool was no longer necessary
to include, neither is that variable.  Silences a warning.
Signed-off-by: Stephen Hutchinson <qyot27@gmail.com>

Also stop rewriting pts.

Check for failure of avformat_new_stream() and propagate
the error code.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

It need not be writable at all. Instead, use temporary buffers
for decorrelation.
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

It is unnecessary and unchecked; the intention seems to be to ensure
that the frame's data is writable, but it does not provide this.
This will be fixed in a latter commit.
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

A decoder is only opened if there is a decoder for the codec,
so every AVCodecContext here has AVCodecContext.codec set.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

This is a prerequisite to continue using the decoder at all
to decode the next interval (if any).
This fixes a regression introduced in commit
2a88ebd0 and reported in ticket #8657.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

Also stop using audio fifo.

Also stop using fifo and excessive peeking.

This is not needed at all.

Add support for hevc_qsv to input RGB format frame. It will
transform frame to yuv inside MediaSDK instead of using auto
scale. Now hevc_qsv supports directly encoding BGRA and X2RGB10
format. The X2RGB10 correspond to the A2RGB20 format and BGRA
correspond to RGB4 format in MediaSDK.
Signed-off-by: Wenbin Chen <wenbin.chen@intel.com>
Signed-off-by: Haihao Xiang <haihao.xiang@intel.com>

MSDK recognizes both yuv420p10 and yuv420p9 as MFX_FOURCC_P010, but
parameters are different. When decode yuv420p9 video, ffmpeg-qsv will use
yuv420p10le to configure surface which is different with param from
DecoderHeader and this will lead to error. Now change it use
param from decoderHeader to configure surface.
Signed-off-by: Wenbin Chen <wenbin.chen@intel.com>
Signed-off-by: Haihao Xiang <haihao.xiang@intel.com>

Whatever.

This commit added a sizeV option, integrated some identical operations
to a separate function, and updated the CGS for horizontal and vertical
respectively.

The following command is on how to apply sizeV option:

ffmpeg -init_hw_device vulkan -i input.264 -vf \
hwupload,gblur_vulkan=size=127:sigma=20:sizeV=3:sigmaV=0.5,hwdownload,format=yuv420p \
-y out.264
Signed-off-by: Wu Jianhua <jianhua.wu@intel.com>

Signed-off-by: Wu Jianhua <jianhua.wu@intel.com>

Use the commands below to test: (href: https://trac.ffmpeg.org/wiki/Blend)

I. make an image for test
ffmpeg -f lavfi -i color=s=256x256,geq=r='H-1-Y':g='H-1-Y':b='H-1-Y' -frames 1 \
-y -pix_fmt yuv420p test.jpg

II. blend in sw
ffmpeg -i test.jpg -vf "split[a][b];[b]transpose[b];[a][b]blend=all_mode=multiply,\
pseudocolor=preset=turbo" -y multiply_sw.jpg

III. blend in vulkan
ffmpeg -init_hw_device vulkan -i test.jpg -vf "split[a][b];[b]transpose[b];\
[a]hwupload[a];[b]hwupload[b];[a][b]blend_vulkan=all_mode=multiply,hwdownload,\
format=yuv420p,pseudocolor=preset=turbo" -y multiply_vulkan.jpg
Signed-off-by: Wu Jianhua <jianhua.wu@intel.com>

mips has several headers that are only meant for inclusion in another
non-arch specific file; they do not even try to be standalone. So don't
test them in checkheaders.

Also fix vp9dsp_mips.h, an ordinary header missing some includes.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

Fixes checkheaders on aarch64.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

Fixes make checkheaders on PPC, for which no arch-specific header
exists that indirectly includes attributes.h.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

Only include it if it is needed, namely if __MMX__ is undefined.

X86 is currently the only arch where lavu/cpu.h is basically
automatically included (for internal development): #if ARCH_X86
is true, lavu/internal.h (which is basically included everywhere)
includes lavu/x86/emms.h which can mask missing inclusions
of lavu/cpu.h if the developer works on x86/x64. This has happened
in 8e825ec3 and also earlier
(see 6d236588).
By including said header only if necessary ordinary developer machines
will behave like non-x86 arches, so that missing inclusions of cpu.h
won't go unnoticed any more.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>